Weight loss in acquired immunodeficiency syndrome (AIDS) is characterized by disproportionately greater loss of lean body mass, portends poor clinical outcome, and leads to debility and impaired quality of life. Increased energy intake along does not consistently increase lean body mass. This project will assess the effectiveness of 2 strategies for increasing lean body mass, namely, testosterone and mass, muscle size and strength in other hypogonadal and frail states. Specifically, this study will examine if testosterone replacement and resistance training when administered in association with an optimized protein and energy intake, will increase lean body mass, muscle size and strength (by increasing protein synthesis) and improve quality of life in HIV infected men who are hypogonadal (serum testosterone levels less than or equal to 300 ng/dL), have lost at least 10% of their usual premorbid weight in the preceding six months, and are free of detectable gastrointestinal dysfunction or acute illness. Subjects will be randomly assigned to one of four groups; testosterone (6 mg/day) alone; diluent alone; resistance exercise plus diluent; testosterone plus resistance exercise. The following outcome measures will be studied before and after 16 weeks of treatment; body weight, lean body mass by DEXA scan and doubly labeled water; muscle size by MRI scan of arm and leg; total energy expenditure by doubly labeled water; protein metabolism by overall nitrogen balance on a defined dietary protein intake, [1-13C]leucine turnover, fractional muscle protein synthesis during [1-13C]leucine infusion, and nitrogen kinetics using [15N] glycine infusion; muscle strength by both effort dependent (1RM) and effort independent (force:EMG method) methods. In addition, markers of testosterone bioavailability, immunological function, and safety parameters will be monitored throughout the study. Quality of life and sexual function will be assessed before and during treatment. Careful selection of target patient population, use of two (testosterone and resistance exercise) interventions previously shown to have anabolic effects in other frail and/or hypogonadal states, state-of- the-art methods, and consideration of power and effect size should maximize the chance of identifying one or more effective strategies for this important public health problem.